Symmetric task

Reading libraries and parameters

library(tidyverse)
library(quickpsy)
library(cowplot)
list.files("R", full.names = TRUE) %>% walk(source)
source("graphical_parameters.R")
source("parameters.R")
load(file = "logdata/dat_sym.RData")

No two guess

fun_sym_no_two_guess <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[5] + + (1 - p[5] - p[6]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[7] + (1 - p[7] - p[8]) * pnorm(x, p[1] - p[2], p[4]))))
fit_sym_no_two_guess <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_no_two_guess,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_no_two_guess$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_no_two_guess$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

Two guess

fun_sym_two_guess <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[5] + p[7] + (1 - 2 * p[5] - p[7]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[6] + (1 - 2 * p[6] - p[7]) * pnorm(x, p[1] - p[2], p[4]))))
fit_sym_two_guess <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_two_guess,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_two_guess$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_two_guess$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess vs two guess

sym_no_two_guess_vs_two_guess <- model_selection_lrt(
  fit_sym_no_two_guess$logliks, 
  fit_sym_two_guess$logliks) 
sym_no_two_guess_vs_two_guess %>% 
  group_by(best) %>% 
  count()

best_sym_no_two_guess <- sym_no_two_guess_vs_two_guess %>% 
  filter(best == "first") %>% 
  select(subject, vertical)
best_sym_two_guess <- sym_no_two_guess_vs_two_guess %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

No two guess same slope

fun_sym_no_two_guess_same_slope <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[4] + + (1 - p[4] - p[5]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[6] + (1 - p[6] - p[7]) * pnorm(x, p[1] - p[2], p[3]))))
fit_sym_no_two_guess_same_slope <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_no_two_guess_same_slope,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_no_two_guess_same_slope$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_no_two_guess_same_slope$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess vs no two guess same slope

sym_no_two_guess_vs_no_two_guess_same_slope <- model_selection_lrt(
  fit_sym_no_two_guess$logliks, 
  fit_sym_no_two_guess_same_slope$logliks) 
sym_no_two_guess_vs_no_two_guess_same_slope %>%
  semi_join(best_sym_no_two_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_no_two_guess_no_same_slope <- sym_no_two_guess_vs_no_two_guess_same_slope %>% 
  semi_join(best_sym_no_two_guess) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_two_guess_same_slope <- sym_no_two_guess_vs_no_two_guess_same_slope %>% 
  semi_join(best_sym_no_two_guess) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

Sym guess

fun_sym_guess <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[5] + (1 - 2 * p[5]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[6] + (1 - 2 * p[6]) * pnorm(x, p[1] - p[2], p[4]))))
fit_sym_guess <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_guess,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, pini_scale, 
                              pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_guess$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_guess$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Sym guess vs no sym guess

sym_two_guess_vs_sym_guess <- model_selection_lrt(
  fit_sym_two_guess$logliks, 
  fit_sym_guess$logliks) 
sym_two_guess_vs_sym_guess %>%
  semi_join(best_sym_two_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_guess <- best_sym_two_guess

Sym same guess

fun_sym_same_guess <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[5] + (1 - 2 * p[5]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[5] + (1 - 2 * p[5]) * pnorm(x, p[1] - p[2], p[4]))))
fit_sym_same_guess <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess vs no same slope

sym_same_guess_vs_no_same_guess <- model_selection_lrt(
  fit_sym_guess$logliks, 
  fit_sym_same_guess$logliks) 
sym_same_guess_vs_no_same_guess %>%
  semi_join(best_sym_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_no_same_guess <- sym_same_guess_vs_no_same_guess %>% 
  semi_join(best_sym_guess) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

best_sym_same_guess <- sym_same_guess_vs_no_same_guess %>% 
  semi_join(best_sym_guess) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

No same guess same slope

fun_sym_guess_same_slope <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[5] + (1 - 2 * p[5]) * pnorm(x, p[1] - p[2], p[3]))))
fit_sym_guess_same_slope <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_guess_same_slope,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, 
                              pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_guess_same_slope$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_guess_same_slope$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

No same guess vs no same guess same slope

sym_same_guess_vs_same_guess_same_slope <- model_selection_lrt(
  fit_sym_guess$logliks, 
  fit_sym_guess_same_slope$logliks) 
sym_same_guess_vs_same_guess_same_slope %>%
  semi_join(best_sym_no_same_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_same_guess_same_slope <- best_sym_no_same_guess
  

Absent lapses

fun_sym_absent_lapses <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1] - p[2], p[4]))))
fit_sym_absent_lapses <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses vs no absent lapses

sym_absent_lapses_vs_no_absent_lapses <- model_selection_lrt(
  fit_sym_same_guess$logliks, 
  fit_sym_absent_lapses$logliks) 
sym_absent_lapses_vs_no_absent_lapses %>%
  semi_join(best_sym_same_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_no_absent_lapses <- sym_absent_lapses_vs_no_absent_lapses %>% 
  semi_join(best_sym_same_guess) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

best_sym_absent_lapses <- sym_absent_lapses_vs_no_absent_lapses %>% 
  semi_join(best_sym_same_guess) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

Sym same guess same slope

fun_sym_same_guess_same_slope <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1] - p[2], p[3]))))
fit_sym_same_guess_same_slope <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_same_slope,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_same_slope$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_same_slope$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

No absent lapses vs no absent lapses same slope

sym_no_absent_lapses_vs_no_absent_lapses_same_slope <- model_selection_lrt(
  fit_sym_same_guess$logliks, 
  fit_sym_same_guess_same_slope$logliks) 
sym_no_absent_lapses_vs_no_absent_lapses_same_slope %>%
  semi_join(best_sym_no_absent_lapses) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_absent_lapses_no_same_slope <- sym_no_absent_lapses_vs_no_absent_lapses_same_slope %>% 
  semi_join(best_sym_no_absent_lapses) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_absent_lapses_same_slope <- sym_no_absent_lapses_vs_no_absent_lapses_same_slope %>% 
  semi_join(best_sym_no_absent_lapses) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

Absent lapses same slope

fun_sym_absent_lapses_same_slope <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1] + p[2], p[3]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1] - p[2], p[3]))))
fit_sym_absent_lapses_same_slope <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_same_slope,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_origin, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_same_slope$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_same_slope$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses vs absent lapses same slope

sym_absent_lapses_vs_absent_lapses_same_slope <- model_selection_lrt(
  fit_sym_absent_lapses$logliks, 
  fit_sym_absent_lapses_same_slope$logliks) 
sym_absent_lapses_vs_absent_lapses_same_slope %>%
  semi_join(best_sym_absent_lapses) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_absent_lapses_no_same_slope <- sym_absent_lapses_vs_absent_lapses_same_slope %>% 
  semi_join(best_sym_absent_lapses) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_absent_lapses_same_slope <- sym_absent_lapses_vs_absent_lapses_same_slope %>% 
  semi_join(best_sym_absent_lapses) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

Averages, curves and parameters (checking)

sym_averages_s_vs_d_test <- 
  (fit_sym_no_two_guess_same_slope$averages %>% semi_join(best_sym_no_two_guess_same_slope))

Joining, by = c("subject", "vertical")

sym_curves_s_vs_d_test <- 
  (fit_sym_no_two_guess_same_slope$curves %>% semi_join(best_sym_no_two_guess_same_slope))

Joining, by = c("subject", "vertical")

ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = sym_averages_s_vs_d_test, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = sym_curves_s_vs_d_test, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

sym_averages_s_vs_d_test <- 
  (fit_sym_same_guess_same_slope$averages %>% semi_join(best_sym_no_absent_lapses_same_slope))

Joining, by = c("subject", "vertical")

sym_curves_s_vs_d_test <- 
  (fit_sym_same_guess_same_slope$curves %>% semi_join(best_sym_no_absent_lapses_same_slope))

Joining, by = c("subject", "vertical")

ggplot() + facet_wrap(subject ~ vertical) +
  geom_point(data = sym_averages_s_vs_d_test, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = sym_curves_s_vs_d_test, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

sym_averages_s_vs_d_test <- 
  (fit_sym_absent_lapses$averages %>% semi_join(best_sym_absent_lapses_no_same_slope))

Joining, by = c("subject", "vertical")

sym_curves_s_vs_d_test <- 
  (fit_sym_absent_lapses$curves %>% semi_join(best_sym_absent_lapses_no_same_slope))

Joining, by = c("subject", "vertical")

ggplot() + facet_wrap(subject ~ vertical) +
  geom_point(data = sym_averages_s_vs_d_test, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = sym_curves_s_vs_d_test, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

Indecision model

Averages, curves and parameters

Plotting thresholds

ggplot() + facet_wrap(subject ~ vertical) +
  geom_point(data = sym_averages_s_vs_d, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = sym_curves_s_vs_d, 
            aes(x = x, y = y, color = references)) +
  geom_segment(data = sym_thre_s_vs_d, 
           aes(x = thre, xend = thre, y = 0 , yend = prob, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess same slope zero

fun_sym_no_two_guess_same_slope_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[2] + + (1 - p[2] - p[3]) * pnorm(x, 0, p[1]), 
    function(x, p) p[4] + (1 - p[4] - p[5]) * pnorm(x, 0, p[1]))))
fit_sym_no_two_guess_same_slope_zero <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_no_two_guess_same_slope_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_no_two_guess_same_slope_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_no_two_guess_same_slope_zero$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess same slope zero vs no two guess same slope

sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_zero <- model_selection_lrt(
  fit_sym_no_two_guess_same_slope$logliks, 
  fit_sym_no_two_guess_same_slope_zero$logliks) 
sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_zero %>%
  semi_join(best_sym_no_two_guess_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_no_two_guess_same_slope_no_zero <- sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_zero %>% 
  semi_join(best_sym_no_two_guess_same_slope) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_two_guess_same_slope_zero <- sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_zero %>% 
  semi_join(best_sym_no_two_guess_same_slope) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "zero")

Joining, by = c("subject", "vertical")

No two guess same slope s

fun_sym_no_two_guess_same_slope_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + + (1 - p[3] - p[4]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[5] + (1 - p[5] - p[6]) * pnorm(x, p[1], p[2]))))
fit_sym_no_two_guess_same_slope_s <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_no_two_guess_same_slope_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_no_two_guess_same_slope_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_no_two_guess_same_slope_s$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess same slope vs no two guess same slope s

sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_s <- model_selection_lrt(
  fit_sym_no_two_guess_same_slope$logliks, 
  fit_sym_no_two_guess_same_slope_s$logliks) 
sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_s %>%
  semi_join(best_sym_no_two_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

No two guess same slope d

fun_sym_no_two_guess_same_slope_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + + (1 - p[3] - p[4]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[5] + (1 - p[5] - p[6]) * pnorm(x, -p[1], p[2]))))
fit_sym_no_two_guess_same_slope_d <- quickpsy(dat_sym, orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_no_two_guess_same_slope_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse, pini_lapse, pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_no_two_guess_same_slope_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_no_two_guess_same_slope_d$curves, 
            aes(x = x, y = y, color = references)) +
  theme_grey() + theme(legend.position = "top") 

No two guess same slope vs no two guess same slope d

sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_d <- model_selection_lrt(
  fit_sym_no_two_guess_same_slope$logliks, 
  fit_sym_no_two_guess_same_slope_d$logliks) 
sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_d %>%
  semi_join(best_sym_no_two_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_no_two_guess_same_slope_no_zero_s <- sym_no_two_guess_same_slope_vs_no_two_guess_same_slope_d %>% 
  semi_join(best_sym_no_two_guess_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

No same guess same slope zero

fun_sym_guess_same_slope_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[2] + (1 - 2 * p[2]) * pnorm(x, 0, p[1]), 
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, 0, p[1]))))
fit_sym_guess_same_slope_zero <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_guess_same_slope_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale, 
                              pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_guess_same_slope_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_guess_same_slope_zero$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

No same guess same slope vs no same guess same slope zero

sym_guess_same_slope_vs_sym_guess_same_slope_zero <- model_selection_lrt(
  fit_sym_guess_same_slope$logliks, 
  fit_sym_guess_same_slope_zero$logliks) 
sym_guess_same_slope_vs_sym_guess_same_slope_zero %>%
  semi_join(best_sym_no_same_guess) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_guess_same_slope_no_zero <- sym_guess_same_slope_vs_sym_guess_same_slope_zero %>% 
  semi_join(best_sym_no_same_guess) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_guess_same_slope_zero <- sym_guess_same_slope_vs_sym_guess_same_slope_zero %>% 
  semi_join(best_sym_no_same_guess) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "zero")

Joining, by = c("subject", "vertical")

No same guess same slope s

fun_sym_guess_same_slope_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1], p[2]))))
fit_sym_guess_same_slope_s <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_guess_same_slope_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_guess_same_slope_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_guess_same_slope_s$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

No same guess same slope no zero vs no same guess same slope no zero s

sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_s <- model_selection_lrt(
  fit_sym_guess_same_slope$logliks, 
  fit_sym_guess_same_slope_s$logliks) 
sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_s %>%
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_guess_same_slope_no_zero_no_s <- sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_s %>% 
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

best_sym_guess_same_slope_no_zero_s <- sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_s %>% 
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

No same guess same slope d

fun_sym_guess_same_slope_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, -p[1], p[2]))))
fit_sym_guess_same_slope_d <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_guess_same_slope_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse, pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_guess_same_slope_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_guess_same_slope_d$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

No same guess same slope no zero vs no same guess same slope no zero d

sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_d <- model_selection_lrt(
  fit_sym_guess_same_slope$logliks, 
  fit_sym_guess_same_slope_d$logliks) 
sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_d %>%
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_guess_same_slope_no_zero_no_d <- sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_d %>% 
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_guess_same_slope_no_zero_d <- sym_guess_same_slope_no_zero_vs_guess_same_slope_no_zero_d %>% 
  semi_join(best_sym_guess_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "decision")

Joining, by = c("subject", "vertical")

Sym same guess no same slope

fun_sym_same_guess_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, 0, p[1]), 
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, 0, p[2]))))
fit_sym_same_guess_zero <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_zero$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess no same slope no zero vs Same guess no same slope zero

sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_zero <- model_selection_lrt(
  fit_sym_same_guess$logliks, 
  fit_sym_same_guess_zero$logliks) 
sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_zero %>%
  semi_join(best_sym_no_absent_lapses_no_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

Sym same guess no same slope no zero s

fun_sym_same_guess_no_zero_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1], p[3]))))
fit_sym_same_guess_no_zero_s <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_no_zero_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_no_zero_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_no_zero_s$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess no same slope no zero vs Same guess no same slope no zero s

sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_s <- model_selection_lrt(
  fit_sym_same_guess$logliks, 
  fit_sym_same_guess_no_zero_s$logliks) 
sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_s %>%
  semi_join(best_sym_no_absent_lapses_no_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

Sym same guess no same slope no zero d

fun_sym_same_guess_no_zero_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[4] + (1 - 2 * p[4]) * pnorm(x, -p[1], p[3]))))
fit_sym_same_guess_no_zero_d <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_no_zero_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_no_zero_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_no_zero_d$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess no same slope no zero vs Same guess no same slope no zero d

sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_d <- model_selection_lrt(
  fit_sym_same_guess$logliks, 
  fit_sym_same_guess_no_zero_d$logliks) 
sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_d %>%
  semi_join(best_sym_no_absent_lapses_no_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_same_guess_no_same_slope_no_zero_full <- sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_s %>% 
  semi_join(best_sym_no_absent_lapses_no_same_slope) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) %>% 
  mutate(best = "full")

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_same_guess_no_same_slope_no_zero_s <- sym_same_guess_no_same_slope_no_zero_vs_same_guess_no_same_slope_no_zero_s %>% 
  semi_join(best_sym_no_absent_lapses_no_same_slope) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

Sym same guess same slope zero

fun_sym_same_guess_same_slope_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[2] + (1 - 2 * p[2]) * pnorm(x, 0, p[1]), 
    function(x, p) p[2] + (1 - 2 * p[2]) * pnorm(x, 0, p[1]))))
fit_sym_same_guess_same_slope_zero <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_same_slope_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_same_slope_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_same_slope_zero$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess same slope no zero vs same guess same slope zero

sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_zero <- model_selection_lrt(
  fit_sym_same_guess_same_slope$logliks, 
  fit_sym_same_guess_same_slope_zero$logliks) 
sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_zero %>%
  semi_join(best_sym_no_absent_lapses_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_sym_same_guess_same_slope_no_zero <- sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_zero %>% 
  semi_join(best_sym_no_absent_lapses_same_slope) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) 

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_same_guess_same_slope_zero <- sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_zero %>% 
  semi_join(best_sym_no_absent_lapses_same_slope) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "zero")

Joining, by = c("subject", "vertical")

Sym same guess same slope no zero s

fun_sym_same_guess_same_slope_no_zero_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, p[1], p[2]))))
fit_sym_same_guess_same_slope_no_zero_s <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_same_slope_no_zero_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_same_slope_no_zero_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_same_slope_no_zero_s$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess same slope no zero vs same guess same slope no zero s

sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_s <- model_selection_lrt(
  fit_sym_same_guess_same_slope$logliks, 
  fit_sym_same_guess_same_slope_no_zero_s$logliks) 
sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_s %>%
  semi_join(best_sym_same_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

Sym same guess same slope no zero d

fun_sym_same_guess_same_slope_no_zero_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, p[1], p[2]), 
    function(x, p) p[3] + (1 - 2 * p[3]) * pnorm(x, -p[1], p[2]))))
fit_sym_same_guess_same_slope_no_zero_d <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_same_guess_same_slope_no_zero_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, 
                              pini_lapse),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_same_guess_same_slope_no_zero_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_same_guess_same_slope_no_zero_d$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Same guess same slope no zero vs same guess same slope no zero d

sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_d <- model_selection_lrt(
  fit_sym_same_guess_same_slope$logliks, 
  fit_sym_same_guess_same_slope_no_zero_d$logliks) 
sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_d %>%
  semi_join(best_sym_same_guess_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_same_guess_same_slope_no_zero_full <- sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_s %>% 
  semi_join(best_sym_same_guess_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) %>% 
  mutate(best = "full")

Joining, by = c("subject", "vertical")

### Add to s vs d
best_sym_same_guess_same_slope_no_zero_s <- sym_same_guess_same_slope_no_zero_vs_same_guess_same_slope_no_zero_s %>% 
  semi_join(best_sym_same_guess_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

Absent lapses no same slope zero

fun_sym_absent_lapses_no_same_slope_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, 0, p[1]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, 0, p[2]))))
fit_sym_absent_lapses_no_same_slope_zero <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_no_same_slope_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_no_same_slope_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_no_same_slope_zero$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses no same slope vs absent lapses no same slope zero

rr sym_absent_lapses_no_same_slope_vs_absent_lapses_no_same_slope_zero <- model_selection_lrt( fit_sym_absent_lapses\(logliks, fit_sym_absent_lapses_no_same_slope_zero\)logliks) sym_absent_lapses_no_same_slope_vs_absent_lapses_no_same_slope_zero %>% semi_join(best_sym_absent_lapses_no_same_slope) %>% group_by(best) %>% count()

Joining, by = c(\subject\, \vertical\)

Absent lapses no same slope no zero s

fun_sym_absent_lapses_no_same_slope_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[2]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[3]))))
fit_sym_absent_lapses_no_same_slope_s <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_no_same_slope_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_no_same_slope_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_no_same_slope_s$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses no same slope no zero vs absent lapses no same slope no zero s

sym_absent_lapses_no_same_slope_no_zero_vs_absent_lapses_no_same_slope_no_zero_s <- model_selection_lrt(
  fit_sym_absent_lapses$logliks, 
  fit_sym_absent_lapses_no_same_slope_s$logliks) 
sym_absent_lapses_no_same_slope_no_zero_vs_absent_lapses_no_same_slope_no_zero_s %>%
  semi_join(best_sym_absent_lapses_no_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

Absent lapses no same slope no zero s

fun_sym_absent_lapses_no_same_slope_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[2]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, -p[1], p[3]))))
fit_sym_absent_lapses_no_same_slope_d <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_no_same_slope_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_no_same_slope_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_no_same_slope_d$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses no same slope no zero vs absent lapses no same slope no zero d

sym_absent_lapses_no_same_slope_no_zero_vs_absent_lapses_no_same_slope_no_zero_d <- model_selection_lrt(
  fit_sym_absent_lapses$logliks, 
  fit_sym_absent_lapses_no_same_slope_d$logliks) 
sym_absent_lapses_no_same_slope_no_zero_vs_absent_lapses_no_same_slope_no_zero_d %>%
  semi_join(best_sym_absent_lapses_no_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_absent_lapses_no_same_slope_no_zero_s <- sym_absent_lapses_no_same_slope_no_zero_vs_absent_lapses_no_same_slope_no_zero_d %>% 
  semi_join(best_sym_absent_lapses_no_same_slope) %>% 
  filter(best == "first") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

Absent lapses same slope zero

fun_sym_absent_lapses_same_slope_zero <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, 0, p[1]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, 0, p[1]))))
fit_sym_absent_lapses_same_slope_zero <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_same_slope_zero,
                xmin = -3, xmax = 3,
                parini = list(pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_same_slope_zero$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_same_slope_zero$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses same slope no zero vs absent lapses same slope zero

sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_zero <- model_selection_lrt(
  fit_sym_absent_lapses_same_slope$logliks, 
  fit_sym_absent_lapses_same_slope_zero$logliks) 
sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_zero %>%
  semi_join(best_sym_absent_lapses_same_slope) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_absent_lapses_same_slope_no_zero <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_zero %>% 
  semi_join(best_sym_absent_lapses_same_slope) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

### Add to s vs d
best_absent_lapses_same_slope_zero <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_zero %>% 
  semi_join(best_sym_absent_lapses_same_slope) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "zero")

Joining, by = c("subject", "vertical")

Absent lapses same slope no zero s

fun_sym_absent_lapses_same_slope_no_zero_s <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[2]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[2]))))
fit_sym_absent_lapses_same_slope_no_zero_s <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_same_slope_no_zero_s,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_same_slope_no_zero_s$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_same_slope_no_zero_s$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses same slope no zero vs absent lapses same slope no zero s

sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s <- model_selection_lrt(
  fit_sym_absent_lapses_same_slope$logliks, 
  fit_sym_absent_lapses_same_slope_no_zero_s$logliks) 
sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s %>%
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

best_absent_lapses_same_slope_no_zero_no_s <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s %>% 
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  select(subject, vertical)

Joining, by = c("subject", "vertical")

best_absent_lapses_same_slope_no_zero_s <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s %>% 
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) 

Joining, by = c("subject", "vertical")

Absent lapses same slope no zero d

fun_sym_absent_lapses_same_slope_no_zero_d <-  dat_sym %>% 
  distinct(references) %>% 
  bind_cols(tibble(fun = c(
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, p[1], p[2]), 
    function(x, p) 0.01 + (1 - 2 * 0.01) * pnorm(x, -p[1], p[2]))))
fit_sym_absent_lapses_same_slope_no_zero_d <- quickpsy(dat_sym, 
                               orientation, response, 
                grouping = .(subject, vertical, references),
                fun = fun_sym_absent_lapses_same_slope_no_zero_d,
                xmin = -3, xmax = 3,
                parini = list(pini_origin, pini_scale),
                bootstrap = "none")
ggplot() + facet_wrap(subject ~ vertical, ncol = 6) +
  geom_point(data = fit_sym_absent_lapses_same_slope_no_zero_d$averages, 
             aes(x = orientation, y = prob, color = references)) +
  geom_line(data = fit_sym_absent_lapses_same_slope_no_zero_d$curves, 
            aes(x = x, y = y, color = references, lty = "all")) +
  theme_grey() + theme(legend.position = "top") 

Absent lapses same slope no zero vs absent lapses same slope no zero d

sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_d <- model_selection_lrt(
  fit_sym_absent_lapses_same_slope$logliks, 
  fit_sym_absent_lapses_same_slope_no_zero_d$logliks) 
sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_d %>%
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  group_by(best) %>% 
  count()

Joining, by = c("subject", "vertical")

### Add to s vs d
best_absent_lapses_same_slope_no_zero_d <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_d %>% 
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "decision")

Joining, by = c("subject", "vertical")

### Add to s vs d
best_absent_lapses_same_slope_no_zero_no_d_s <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s %>% 
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  filter(best == "second") %>% 
  select(subject, vertical) %>% 
  mutate(best = "sensory")

Joining, by = c("subject", "vertical")

### Add to s vs d
best_absent_lapses_same_slope_no_zero_no_s <- sym_absent_lapses_same_slope_no_zero_vs_absent_lapses_same_slope_no_zero_s %>% 
  semi_join(best_absent_lapses_same_slope_no_zero) %>% 
  filter(best == "first") %>% 
  filter(!(subject == 9 & vertical == FALSE)) %>% 
  select(subject, vertical) %>% 
  mutate(best = "full")

Joining, by = c("subject", "vertical")

Add all best

Save data

save(sym_averages_s_vs_d_best, file = "logdata/sym_averages_s_vs_d_best.RData")
save(sym_curves_s_vs_d_best, file = "logdata/sym_curves_s_vs_d_best.RData")
save(sym_par_s_vs_d_best, file = "logdata/sym_par_s_vs_d_best.RData")
save(sym_par_s_vs_d_best_long, file = "logdata/sym_par_s_vs_d_best_long.RData")
save(sym_par_s_vs_d_best_abs, file = "logdata/sym_par_s_vs_d_best_abs.RData")